Re: A proposal for specifying the 'view-source' URI scheme

Part of the reason for introducing the *provisional* scheme registry was to 
provide a single point where developers could find out about known URI schemes 
deployed or under consideration, irrespective of how good or bad they may be. 
The other option, which we'd like to avoid, is to have lots of invented schemes 
in the wild that developers are not aware of, which in turn leads to potentially 
conflicting uses of a given scheme name.

And in some cases, they might even be registered with an exhortation like "DO 
NOT USE THIS".

#g
--

Eric Johnson wrote:
> I guess my take on it, which seems to align with Martin's, is that 
> whatever intent behind the view-source URI scheme, it appears to solve 
> the problem in the wrong way. The functionality can be handled at the 
> presentation/application layer in the browser, not at the extremely 
> low-level of a URI scheme. And of course, if you want to view source, 
> currently you can simply do Ctrl+U, or your browser's equivalent.
> 
> If it is flawed, then perhaps it is sufficient to have a Wikipedia page, 
> and not have it documented at the IETF?
> 
> -Eric.
> 
> On 4/15/11 6:02 AM, Graham Klyne wrote:
>> Martin,
>>
>> I'm guessing your comments are missing Mykyta's point.  Without 
>> knowing all the  details, I'm guessing that a view source URI scheme 
>> *has been implemented* as a mechanism for displaying page source (cf. 
>> http://en.wikipedia.org/wiki/View-source_URI_scheme), and Mykyta is 
>> asking if it should be documented and registered as provisional.
>>
>> This is separate from any issue of its desirability or recommendation 
>> that it be used.
>>
>> I don't have any especially strong feeling about this.  On balance, I 
>> think having its existence documented is not a bad thing, but not an 
>> activity I'd be inclined to invest a lot of effort in.  Maybe just a 
>> registration template with a link to the Wikipedia page?
>>
>> #g
>> -- 
>>
>>
>>
>> Peter Saint-Andre wrote:
>>> On 4/13/11 12:28 AM, "Martin J. Dürst" wrote:
>>>>
>>>> On 2011/04/12 6:59, Eric Johnson wrote:
>>>>> How fascinating.
>>>>>
>>>>> It strikes me that the point of this URL is to signal to a browser 
>>>>> that
>>>>> it should serve content available via HTTP (and others?) in a 
>>>>> particular
>>>>> presentation. Logically, then, this really should fall to the
>>>>> presentation layer (HTML).
>>>> Well, you can of course use HTML to present source code, by e.g. 
>>>> putting
>>>> it into a <pre> element and escaping '<' and friends.
>>>>
>>>> But the official way to serve source as source is to give it a content
>>>> type of text/plain. The problem with that is that some if not most
>>>> browsers, because of the past dynamics between badly setup servers and
>>>> bugwards compatible browsers, tend to ignore the text/plain media type
>>>> under certain  circumstances. For details, please see
>>>> http://tools.ietf.org/html/draft-abarth-mime-sniff.
>>>>
>>>>> Why use a URI to accomplish that? In HTML, there could just as 
>>>>> easily be
>>>>> an attribute on an <a> tag. If I right-click a link, the browser could
>>>>> easily just show me a "browse-source" option.
>>>> Or the user could press cntl-u for 'view source'.
>>>
>>> That was my thought, too.
>>>
>>>>> Some URIs aren't resources in the sense of HTTP - for example, "tel:",
>>>>> "email:", the one I've been working on ("jms:"), etc - so this doesn't
>>>>> seem like the view-source can be generally prefixed to just anything.
>>>> Strictly speaking, that's not necessary at all. If it makes sense in
>>>> some cases, it may be worth using.
>>>>
>>>>
>>>>> My take: it doesn't make sense to standardize.
>>>> I'd agree to be careful and look at what other browsers do, and check
>>>> with some browser vendors whether they care about it, before spending
>>>> needless standardization cycles.
>>>
>>> +1
>>>
>>> Peter
>>>
>>>
>>>
>>> ------------------------------------------------------------------------
>>>
>>> _______________________________________________
>>> Uri-review mailing list
>>> Uri-review <at> ietf.org
>>> https://www.ietf.org/mailman/listinfo/uri-review
>>
>>
>> _______________________________________________
>> Uri-review mailing list
>> Uri-review <at> ietf.org
>> https://www.ietf.org/mailman/listinfo/uri-review
> _______________________________________________
> Uri-review mailing list
> Uri-review <at> ietf.org
> https://www.ietf.org/mailman/listinfo/uri-review
> 

_______________________________________________
Uri-review mailing list
Uri-review <at> ietf.org
https://www.ietf.org/mailman/listinfo/uri-review